(A) Schematic illustration of the experimental approach. (B–C) Representative data for combined SE/QCM-D measurements. Areal protein densities (top) are obtained through optical modeling of SE data (see Materials and methods). QCM-D responses (bottom) are recorded in parallel, and normalized frequency shifts Δfi/i and dissipation shifts ΔDi for a selected overtone (i = 3) displayed here. Film thickness is obtained through viscoelastic modeling of QCM-D data (see Materials and methods). Δf = ΔD = 0 corresponds to the functionalized surface before FG domain grafting. (B) Monitoring of FG domain film formation. Nsp1 was exposed to a silica substrate previously functionalized with an SLB (7% tris-NTA). The final grafting density is 4.9 pmol/cm2. Minor perturbations in Δf and ΔD between 33 and 43 min are due to transient variations in the solution temperature in contact with the QCM-D sensor during the rinsing with buffer, and do not represent changes of the Nsp1 film. (C) Time-resolved data for the titration of NTF2 into this Nsp1 film. The NTF2 solution concentration was increased in 12 steps (0.0025, 0.01, 0.05, 0.1, 0.2, 0.5, 1, 2, 3, 5, 7.5 and 10 μM), and then decreased in 16 steps ((2/3)j × 10 μM, with j = 1, …, 16), followed by continuous rinsing with buffer solution to remove all NTF2 from the bulk solution. The rapid binding and unbinding of NTF2 observed here is representative for all titration measurements performed, and binding equilibriums could thus be readily attained. Moreover, binding of NTF2 was largely reversible, with less than 7% of the maximal binding remaining following rinsing in buffer for any given measurement. For Impβ, more than 75% were readily eluted from 5 pmol/cm2 Nsp1 films, and we had previously shown this NTR to elute close to completely from 10 pmol/cm2 Nsp1 films (Eisele et al., 2012; Eisele et al., 2010).